Venous thromboembolism (VTE), including both pulmonary embolism and deep vein thrombosis, is an age related condition with potentially catastrophic clinical consequences. Although VTE occurs as frequently as stroke in some populations, its epidemiology is not well understood. The broad goal of this application is to clarify the determinants of VTE, with a particular focus on those influenced by age, both to identify modifiable risk factors as well as those people who would benefit most from preventive strategies. Studies will utilize data from five populations: 39,876 women who participated in the Women's Health Study for over 10 years;29,071 men who participated in the Physicians'Health Studies I and II, of whom 22,071 have over 20 years of follow-up since the initiation of the first trial;8,171 women at high risk of cardiovascular disease who participated in the Women's Antioxidant Cardiovascular Disease Study for an average of 8 years, 508 people with prevalent, idiopathic VTE who participated in the PREVENT trial;and 17,802 people from 26 countries currently participating in the JUPITER trial of statins and cardiovascular disease. Parallel, centralized methods were used to review records and confirm reported cases of VTE. Archived blood samples are available for over 75% of participants. Main relationships to be evaluated include: whether intake of whole grains, fruit, and vegetables influence risk of VTE;the influence of comorbid conditions, especially cancer, cardiovascular disease, stroke and diabetes, on risk of VTE;interactions of genetic determinants of VTE, including factor V Leiden, the prothrombin mutation and polymorphisms in the MTHFR gene with age;whether genetic variants in adhesion molecules influence risk of VTE;and the impact of statins and the metabolic syndrome on risk of VTE. The combined populations will include over 1,500 incident cases of VTE. Studies will consider both a first occurrence of VTE as well as the risk of recurrent VTE. Methods of analysis include Kaplan-Meier estimates of cumulative risk and estimates of the relative hazard of VTE based on proportional hazards models. Strengths of the studies include their prospective design, high follow-up rates, detailed and reliable, long-term exposure and outcome information, and availability of blood samples on a high percentage of the population. Results will help identify high-risk older people, and potentially modifiable behaviors that influence risk of